The difference between an observational study and an experimental study is whether or not the variable of interest in the study was introduced intentionally and in a controlled way by the researchers (experimental) or if the variable occurs naturally or outside of the researchers' control (observational).
Does the study look at the impact of a variable outside of the researchers' control?
In observational studies, researchers look at data or make observations about a variable or intervention they are interested in, but do not actually control the distribution of.
Look at the methodology section. Does the study identify participants by looking through existing records or data?
Observational studies often choose participants because they are/were already reported to have a specific condition or because their medical records indicate they have received a specific intervention or treatment.
Look for articles that are described in their title or methodology as:
Would there be huge ethical issues if researchers controlled the intervention or variable of interest?
Observational studies often look at things that it would be unethical or impossible for researchers to distribute, cause, or selectively not distribute. Researchers cannot and should not for example expose participants to known diseases or harmful substances, impregnate participants, deny patients standard care, or change participants' identity or status around race, gender, sexuality, military involvement, ethnicity, disability, etc.
An example:
Researchers look at patients who were previously treated for skin cancer's medical records over twenty years compared to records of patients with no history of skin cancer to see if having skin cancer once makes it more likely that you will have it again.
The researchers (for obvious ethical reasons) can't give patients skin cancer to see if having it once makes it more likely to reoccur.
Does the study look at the effects of a variable or intervention that the researchers intentionally introduce to a specific portion of the study participants?
In experimental studies, researchers deliberately introduce a variable or intervention to some study participants so they can judge its effects compared to those who never received that intervention or received a different intervention.
Does the methodology describe randomizing which participants are in the control group?
This is only possible in an experimental study because in an observational study the researchers have no control over who experiences the intervention or variable of interest.
Look for articles that are described in their titles or methodologies as:
An example:
Researchers split a group of patients who had all recently been treated for skin cancer into two groups. The control group received standard discharge care. The other group received standard care and attended a special "Sun Safety Class" to learn tips about using sunscreen, wearing protective clothing, and having regular screenings. Researchers then followed the two groups for twenty years to see if there was any difference in skin cancer recurrence between participants who received the "Sun Safety Class" intervention compared to participants who did not.
The researchers here deliberately introduce a variable (the "Sun Safety Class") they control to some of the study participants.
The best place to search for Observational Studies is PubMed. PubMed makes it easy by providing an observational studies filter that you can use to limit your results.
You can find this filter by:
This video shows goes over the basics of accessing PubMed and using it to filter for specific article types.